XSpectra · Project

AI-Powered X-Ray Detector That Catches Food Contaminants Other Systems Miss

foodMarket-readyTRL 8

Imagine a security scanner at an airport, but instead of scanning luggage, it scans food on a production line — and it can tell the difference between a piece of bone, a tiny stone, or a sliver of plastic in real time. Regular X-ray detectors see shapes, but XSpectra reads the actual chemical makeup of what's passing through, using deep learning AI to spot contaminants that current technology simply cannot detect. It took over 8 years of R&D to combine advanced photonics, nuclear electronics, and artificial intelligence into one inspection system. Five units have already been installed on real production lines.

By the numbers

Years of R&D behind the technology

XSpectra reference champion units installed on production lines

Core innovations combined (photonics, nuclear electronics, deep learning AI)

Total project deliverables completed

The business problem

What needed solving

Food manufacturers lose significant revenue every year from product recalls and batch discards caused by foreign body contamination that current inspection technology simply cannot detect. Conventional X-ray and metal detectors miss non-metallic contaminants like glass, bone, stone, and certain plastics, leaving producers exposed to safety incidents, regulatory penalties, and wasted product.

The solution

What was built

An industrial-grade, real-time X-ray multi-spectrum inspection system that combines photonics, nuclear electronics, and deep learning AI to detect food contaminants invisible to conventional technology. The project delivered 5 installed reference champion units on production lines and 6 total deliverables.

Audience

Who needs this

Food manufacturers running high-volume packaging lines with contamination riskPharmaceutical companies needing in-line quality inspection for tablets and capsulesMaterial recycling facilities requiring accurate real-time material sortingSecurity screening companies looking for advanced detection capabilitiesNon-destructive testing providers serving aerospace or automotive sectors

Business applications

Who can put this to work

Food Manufacturing

mid-size

Target: Food production companies running high-speed packaging lines

If you are a food manufacturer dealing with product recalls and batch discards due to undetectable foreign bodies — this project built an X-ray inspection system combining 3 high innovations (photonics, nuclear electronics, deep learning AI) that identifies contaminants invisible to conventional detectors. With 5 reference units already installed on production lines, it directly reduces recall costs and minimizes wasted product batches.

Pharmaceutical Manufacturing

enterprise

Target: Pharma companies needing in-line quality inspection

If you are a pharmaceutical manufacturer struggling with contamination detection in tablets, capsules, or powdered products — this project developed a real-time multi-spectrum X-ray system that analyzes physical-chemical features of materials passing through the line. After 8 years of R&D and successful field testing in food production, the same technology applies directly to pharma quality control without stopping the line.

Waste Management & Recycling

mid-size

Target: Material recycling facilities sorting mixed waste streams

If you are a recycling operator facing poor material separation accuracy and contaminated output streams — this project created an inspection detector that reads the chemical composition of materials in real time using deep learning. Originally proven in food safety with 5 installed units, the same capability identifies and classifies materials on recycling conveyor belts to improve sorting purity.

Frequently asked

Quick answers

What does this system cost compared to conventional X-ray inspection?

Based on available project data, specific pricing is not disclosed. However, the system is positioned as reducing costs from product batch discards and expensive product recalls, which suggests the ROI comes from avoided losses rather than upfront savings. The technology combines 3 proprietary innovations built in-house, indicating a premium but self-contained solution.

Can this scale to high-speed industrial production lines?

Yes. The objective confirms that XSpectra completed industrial prototype testing and was installed for field testing on an actual production line. The project delivered 5 reference champion units on real production environments, demonstrating industrial-scale capability with real-time inspection speed.

What is the IP and licensing situation?

XSpectra is developed entirely in-house by XNEXT SPA, an Italian SME that owns all 3 core innovations: photonics, nuclear electronics, and deep learning AI algorithms. As a single-partner project with 100% industry consortium, there are no university or research institute IP co-ownership complications. Licensing terms would be negotiated directly with XNEXT.

Does this meet food safety regulations?

The system is designed to detect foreign bodies that are currently not detectable by existing approved technologies. While specific regulatory certifications are not detailed in the project data, the 5 installed reference units on production lines suggest it operates within existing food safety compliance environments. Contact the coordinator for current certification status.

How long does it take to install and integrate into an existing line?

Based on available project data, the system has been successfully installed on production lines as demonstrated by the 5 reference champion deliverables. Specific installation timelines are not disclosed, but the field-tested, industrial-grade design suggests integration into existing conveyor-based inspection points.

What contaminants can it detect that others cannot?

XSpectra uses X-ray multi-spectrum analysis to read the physical-chemical features of materials in real time, identifying foreign bodies that conventional single-energy X-ray or metal detectors miss. This includes non-metallic contaminants like glass, stone, bone, and certain plastics that lack the density contrast needed by traditional systems.

What is the technology maturity — is this still experimental?

This is not experimental. The project ran for over 8 years of total R&D, completed industrial prototype testing, and delivered 5 installed reference units on production lines. Funded under the SME Instrument Phase 2 scheme, which specifically supports close-to-market commercialization. The project is closed and the product was positioned for worldwide commercial launch.

Consortium

Who built it

This is a single-company project: XNEXT SPA, an Italian SME that is both the sole partner and coordinator. The 100% industry, 100% SME consortium with zero university or research partners is unusual but telling — it means all intellectual property sits with one commercial entity, making licensing and procurement straightforward with no multi-party IP negotiations. The SME Instrument Phase 2 funding validates that the EU assessed this as a commercially viable, close-to-market technology. For a business buyer, this simplifies everything: one company to talk to, one decision-maker, no academic co-owners to align.

How to reach the team

XNEXT SPA is an Italian SME based in Italy. Contact through their website xspectra.eu or request an introduction through SciTransfer.

Order a report on this consortium

Next steps

Talk to the team behind this work.

Want to explore whether XSpectra fits your production line? SciTransfer can arrange a direct introduction to the XNEXT team and help you evaluate the technology for your specific use case.

Order a report on this topic More in Food & Agriculture